Combining both ensemble and dynamic classifier selection schemes for prediction of mobile internet subscribers

Abstract The recent trend of combining internet with cellular phone gives an important chance to mobile telecommunication industry. It is mainly due to the fact that the mobile internet service based on cellular phone is not only more profitable than usual calling service but also it creates new incomes. With this background, mobile telecommunication companies wish to predict the subscribers of mobile internet service from the existing cellular phone users. In this case classification models play important roles and the improvement of classification accuracy has significant target marketing effects due to its huge market size. One of the main efforts made to improve the classification accuracy is a combined classification. This paper proposes a new combined classifier based on both ensemble (bagging) and dynamic classifier selection schemes (DCS). The proposed method is applied to the call data of subscribers and its result confirms the superiority of our approach over bagging and DCS methods.

[1]  Leo Breiman,et al.  Bagging Predictors , 1996, Machine Learning.

[2]  R. Mojena,et al.  Hierarchical Grouping Methods and Stopping Rules: An Evaluation , 1977, Comput. J..

[3]  G. W. Milligan,et al.  An examination of procedures for determining the number of clusters in a data set , 1985 .

[4]  Fabio Roli,et al.  Dynamic classifier selection based on multiple classifier behaviour , 2001, Pattern Recognit..

[5]  Fabio Roli,et al.  An approach to the automatic design of multiple classifier systems , 2001, Pattern Recognit. Lett..

[6]  Jiri Matas,et al.  On Combining Classifiers , 1998, IEEE Trans. Pattern Anal. Mach. Intell..

[7]  Ludmila I. Kuncheva,et al.  Clustering-and-selection model for classifier combination , 2000, KES'2000. Fourth International Conference on Knowledge-Based Intelligent Engineering Systems and Allied Technologies. Proceedings (Cat. No.00TH8516).

[8]  Robert P. W. Duin,et al.  Bagging, Boosting and the Random Subspace Method for Linear Classifiers , 2002, Pattern Analysis & Applications.

[9]  David M. Sebert,et al.  A clustering algorithm for identifying multiple outliers in linear regression , 1998 .

[10]  C. J. Whitaker,et al.  Ten measures of diversity in classifier ensembles: limits for two classifiers , 2001 .

[11]  Thomas G. Dietterich.  An Experimental Comparison of Three Methods for Constructing Ensembles of Decision Trees: Bagging, Boosting, and Randomization , 2000, Machine Learning.

[12]  Kevin W. Bowyer,et al.  Combination of Multiple Classifiers Using Local Accuracy Estimates , 1997, IEEE Trans. Pattern Anal. Mach. Intell..

[13]  J. Neter,et al.  Applied Linear Regression Models , 1983 .

[14]  Tin Kam Ho,et al.  The Random Subspace Method for Constructing Decision Forests , 1998, IEEE Trans. Pattern Anal. Mach. Intell..